The present invention relates to a circuit for digitally computing exposure control information for use in a camera that control exposure automatically, and more particularly to a circuit for digitally computing the information necessary for controlling the shutter speed or aperture in an automatically exposure-controlled camera.
The shutter speed T of a camera, that is, the time during which the shutter is opened, is given as: EQU T = K(1/L)(1/ASA)F (1)
where,
L: scene brightness PA1 Asa: film sensitivity PA1 F: aperture PA1 K: constant of proportionality
In a camera, in other words, the proper shutter speed can be derived from the product of the reciprocal of the scene brightness value (1/L), the reciprocal of the film sensitivity value (1/ASA), and the aperture value F.
The conventional automatically exposure controlled camera has employed an analog circuit for computing the shutter speed. In such a circuit, the shutter speed has had to be computed through the steps of logarithmically compressing values of the respective information, calculating the sum of the compressed values in place of the product of the original values, and logarithmically expanding the result of the calculation, because the ratio of a maximum value to the minimum value for the brightness information exceeds 10.sup.6. Such an analog circuit, however, requires intricate adjustment and uses circuit elements with characteristics of high accuracy and precision, in order to maintain substantial operating stability against voltage and temperature variations. One prior art approach to solve this problem has been to use a digital circuit using semiconductor circuit elements. This circuit may be advantageous in that it can be operated with high accuracy without intricate adjustment and that a shutter speed can be readily displayed. On the other hand, however, the prior art digital circuit has taken an unpractically long time for photometering and computing. The approach to reduce the photometering or computing time by reducing the number of significant figures of the values of the respective information to be computed has led to low accuracy in computation.